This document outlines a research protocol for a forthcoming network meta-analysis. Because it is a protocol (a registered plan for research) rather than a completed study, it describes the intended methodology and objectives rather than final experimental findings.
TL;DR
Researchers are launching a large-scale network meta-analysis to determine which Traditional Chinese Medicine (TCM) preparations, when combined with Western medicine, are most effective at treating mild hand-foot-and-mouth disease (HFMD). By synthesizing data from numerous clinical trials, the study aims to provide a data-driven ranking of integrated treatment schemes to optimize recovery times for pediatric patients.
What Problem Does This Paper Address?
Hand-foot-and-mouth disease (HFMD), often caused by human enteroviruses, is a significant public health concern, particularly among children. While conventional Western medicine (such as antivirals like ribavirin) is the standard of care, Traditional Chinese Medicine is frequently used as a complementary therapy in China.
The challenge lies in the sheer variety of TCM preparations available—ranging from oral herbs to injectable formulas. Clinicians currently lack a clear, evidence-based hierarchy to determine which specific TCM-Western medicine combination yields the best results. Standard meta-analyses typically only compare two treatments at a time. This paper proposes using a network meta-analysis (NMA) to bridge that gap, allowing for the simultaneous comparison and ranking of multiple treatment protocols, even those that have never been compared directly in a single head-to-head trial.
Methods
The authors have designed a rigorous framework to aggregate and analyze existing clinical data:
- Study Selection: The team will exclusively include Randomized Controlled Trials (RCTs) involving patients with mild HFMD.
- The "Network" Approach: Unlike a standard meta-analysis, the NMA uses a Bayesian framework to perform indirect comparisons. If Treatment A is compared to Western Medicine (WM) in one study, and Treatment B is compared to WM in another, the NMA can statistically infer the relative efficacy of A vs. B.
- Computational Tools: The analysis will be performed using the
gemtcpackage in R (version 4.4.3) for Bayesian modeling and Stata 17.0 for frequentist validation. - Statistical Metrics:
- Binary outcomes (e.g., total efficacy rate) will be measured using Odds Ratios (OR).
- Continuous outcomes (e.g., days until fever breaks) will be measured using Mean Difference (MD).
- Uncertainty will be expressed via 95% Credibility Intervals (CrI).
- Quality Control: The researchers will use the Cochrane Risk of Bias tool to assess the internal validity of the included trials, ensuring that poor-quality data does not skew the results.
Datasets
As a meta-analysis, this study does not generate a new primary dataset but instead aggregates data from several major medical databases:
- Chinese Databases: CNKI (China National Knowledge Infrastructure), Wanfang, VIP, and CBM (Chinese Biomedical Literature Database).
- International Databases: PubMed.
The focus on Chinese databases is critical here, as the majority of RCTs involving TCM preparations are published in regional journals.
Key Insights and Ideas
While the final results are pending, the protocol highlights several sophisticated analytical strategies:
- Ranking Treatments: The study intends to go beyond "does it work?" to "which is best?" by calculating the probability of each treatment being the top-ranked intervention for specific symptoms.
- Subgroup Granularity: The researchers plan to differentiate between oral and injectable TCM preparations. This is a vital distinction for clinical practice, as delivery methods can significantly impact patient compliance and physiological absorption.
- Addressing Heterogeneity: By pre-defining a sensitivity analysis (switching between fixed-effect and random-effect models based on data variance), the authors aim to ensure their conclusions remain robust even if the underlying trials vary in quality or patient demographics.
Key Results
Not discussed in the paper. As this is a registered protocol (INPLASY202570108), the data extraction and synthesis phase is currently underway. The primary outcomes to be reported in the future include:
- Total clinical efficacy rate.
- Time required for fever reduction.
- Time required for skin rashes to subside.
- Time required for oral symptoms (like ulcers) to disappear.
Limitations and Future Work
The authors acknowledge (or the protocol implies) several constraints:
- Scope: The study is limited to "mild" cases of HFMD. Results may not generalize to severe or complicated cases involving neurological or pulmonary issues.
- Geographic Bias: Given the nature of TCM, the vast majority of included studies will likely originate from China, which may affect the global generalizability of the findings.
- Standardization: TCM preparations can vary in composition and dosage across different manufacturers; the meta-analysis will have to account for this inherent "noise" in the data.
Future work will involve the execution of this plan and the subsequent publication of the treatment rankings, which could potentially influence pediatric clinical guidelines for HFMD management.
Summary updated